KR102251377B1 - System and method for evaluating interpersonal influence using social data - Google Patents

Abstract

Disclosed is a system and method for evaluating interpersonal influence using social data.
An interpersonal influence evaluation system using social data according to an embodiment of the present invention includes: a social data collection unit that collects social data generated by social network activities of social network service users; An influence direction determination unit for determining an influence direction of a first user by using information about an action subject and an action time of a social network activity obtained from the collected social data; And when it is determined that the direction of influence of the first user is toward the second user, the first user's time interval between the causal action of the first user and the response action of the second user related to the causal action is considered. Including a directional index calculation unit that calculates a directional index on the direction of influence, the direction and intensity of interpersonal influence on various subjects, objects, etc. are evaluated in time series, and the accuracy and reliability of the influence evaluation are improved.

Description

System and method for evaluating interpersonal influence using social data}

The present invention relates to a system and method for evaluating interpersonal influence using social data, and more particularly, to determine the direction and intensity of interpersonal influence on various subjects, targets, etc. using social data generated by social network activity. It relates to a system and method for evaluating impact.

In general, the influence index was used to quantify and evaluate the influence of a specific content based on the citation information of a specific content. Recently, with the development of big data analysis technology, in order to ensure the accuracy and reliability of analysis in various fields, such as market analysis related to specific products, marketing strategy establishment, advertisement model selection, personal profile creation and verification, etc. There is a rapid increase in research that is attempting to expand the influence index evaluation model or develop a new interpersonal influence evaluation model.

However, most of the existing technologies are to calculate the influence index by taking into account the number of reaction contents and the relationship between those who have registered the contents, targeting a specific content. Therefore, it is not possible to expand the application of this influence evaluation method to other targets. Is There is a limit.

In addition, as disclosed in Korean Patent Laid-Open Publication No. 10-2015-0005795, Korean Patent Publication No. 10-1409273, and the like, existing technologies calculate only the intensity index of the influence on the premise that the influence is exerted in only one direction. will be. In other words, since existing technologies cannot measure the direction of influence, there is a problem that they cannot be applied to the evaluation of human-to-human influence for which the direction of influence has not been determined. Furthermore, since existing technologies cannot measure the direction of influence, there is a problem that it is not possible to analyze the flow of influence related to a specific topic, object, etc. in a time series on a social network.

The technical problem to be solved by the present invention is an interpersonal influence evaluation system that evaluates the direction and intensity of interpersonal influence on various subjects, objects, etc. in time series using social data, and improves the accuracy and reliability of the influence evaluation, and Is to provide a way.

In order to solve the above technical problems, the interpersonal influence evaluation system using social data according to an embodiment of the present invention includes a social data collection unit that collects social data generated by social network activities of social network service users. ; An influence direction determination unit for determining an influence direction of a first user by using information about an action subject and an action time of a social network activity obtained from the collected social data; And when it is determined that the direction of influence of the first user is toward the second user, the first user's time interval between the causal action of the first user and the response action of the second user related to the causal action is considered. It includes a directional index calculation unit that calculates a directional index for the direction of influence.

In one embodiment, the system, when it is determined that the influence direction of the first user is toward the second user, the first user's causal action and the reaction action of the second user related to the causal action It may further include an intensity index calculator that calculates an intensity index related to the intensity of the influence of the first user by using the information.

In one embodiment, the influence direction determination unit is, when the second user's action related to the first user's action is later than the first user's action, the influence direction of the first user is determined by the second user. It can be judged as heading.

In an embodiment, the directionality index calculation unit may calculate the directionality index by assigning a weight according to the type of the reaction behavior to a time interval between the causal behavior of the first user and the reaction behavior of the second user. .

In an embodiment, the directionality index calculation unit may calculate the directionality index by further considering the number of reaction actions of the second user.

In one embodiment, the intensity index calculation unit calculates the intensity index in consideration of a time interval between the cause action of the first user and the response action of the second user, the content of the response action, and the number of reaction actions. Can be calculated.

In one embodiment, the intensity index calculation unit, the time interval between the causal action of the first user and the response action of the second user, the content of the response action, and the number of the response action, respectively, the form of the response action The strength index can be calculated by assigning a weight according to.

In an embodiment, the social data collection unit further collects personal communication device usage data generated by at least one of voice communication or text communication through personal communication devices of the social network service users, and the directionality index calculation unit , The directionality index is calculated by further considering communication direction information between the first user and the second user obtained from the collected personal communication device use data, and the intensity index calculation unit includes the collected personal communication device use data The strength index may be calculated by further considering information on the number of communication attempts between the first user and the second user obtained from.

In order to solve the above technical problem, in the method for evaluating interpersonal influence using social data according to an embodiment of the present invention, an influence evaluation system interworking with a network server including a social network service providing server uses social data. A method of evaluating interpersonal influence, comprising: a social data collection step of collecting, by the influence evaluation system, social data generated according to social network activity behaviors of social network service users; An influence direction determination step of determining, by the influence evaluation system, an influence direction of a first user by using information about an action subject of a social network activity and action time information obtained from the collected social data; And when it is determined that the direction of influence of the first user is toward the second user, the influence evaluation system considers a time interval between the causal action of the first user and the reaction action of the second user related to the causal action. And calculating a directional index for the direction of influence of the first user.

In one embodiment, the method comprises, when it is determined that the direction of influence of the first user is directed toward the second user, the influence evaluation system, the cause action of the first user and the second action related to the cause action It may further include an intensity index calculation step of calculating an intensity index related to the intensity of the influence of the first user by using the information on the user's reaction behavior.

According to the present invention, by evaluating influence by using various social data generated by social network activity behavior of social network service users, it is possible to evaluate interpersonal influence on various subjects, objects, and the like.

In particular, it improves the accuracy and reliability of interpersonal influence evaluation by not simply measuring the intensity of influence, but by determining the direction of influence and calculating the influence intensity index in consideration of the determined influence direction.

In addition, it is possible to verify the accuracy and reliability of the determination of the direction of influence by providing a numerical directional index for the determined direction of influence.

In addition, by allowing time-series analysis of the direction and intensity of interpersonal influence in relation to a specific topic, target, etc., influence evaluation technology can be used in various ways such as market analysis, marketing strategy establishment, advertisement model selection, personal profile creation and verification It should be used for a variety of purposes in the field.

Furthermore, those of ordinary skill in the art to which the present invention pertains will clearly understand from the following description that various embodiments according to the present invention can solve various technical problems not mentioned above.

1 is a diagram showing an example of a network environment to which the present invention is applied.
2 is a block diagram showing an interpersonal influence evaluation system using social data according to an embodiment of the present invention.
3 is a flowchart illustrating a method for evaluating interpersonal influence using social data according to an embodiment of the present invention.
4 is a diagram showing an example of an interpersonal influence evaluation result.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to clarify a solution to the technical problem of the present invention. However, in describing the present invention, if the description of the related known technology rather obscure the subject matter of the present invention, a description thereof will be omitted. In addition, terms to be described later are terms defined in consideration of functions in the present invention, and these may be changed according to the intention or custom of the practitioner or the like. Therefore, the definition should be made based on the contents throughout the present specification.

1 shows an example of a network environment to which the present invention is applied.

As shown in FIG. 1, the network environment to which the present invention is applied includes user terminals 10 performing data communication, etc. through a wired/wireless network 30, and network servers 20 providing services to network users. ). The user terminals 10 may include various personal communication devices such as a smart phone 12, a portable computer 14 such as a laptop or notebook capable of wired/wireless data communication, and a desktop PC 16. The network servers 20 may include a social network service providing server 20 and a communication service provider server 24. In general, the social network service providing server 20 is a server that provides a service that establishes an online social network to users who share interests or activities with respect to a specific topic or object. The social network service providing server 20 includes all servers that provide various social network services such as Facebook, Twitter, KakaoTalk, Internet cafes of portal sites, and blogs. Further, the communication company server 24 is a server that provides various communication services such as voice communication and text communication to users of personal communication devices.

The influence evaluation system 100 according to the present invention interworkes with the network servers 20 in such a network environment, and evaluates the influence between people using various social data generated through the user terminals 10. Here, social data includes all kinds of data generated by social network activity behaviors of social network service users. Social network activities include, for example, causal actions such as posting postings and uploading files by service users, and commenting, sharing, embedding, and'approve/disapprove' or'like' by other users on these causal actions. Reactive actions such as input may be included.

Meanwhile, in FIG. 1, the influence evaluation system 100 according to the present invention is shown as a separate system independent from the network servers 20, but is integrated with any one of the network servers 20 according to the embodiment. It can also be composed of.

2 is a block diagram of a system for evaluating interpersonal influence using social data according to an embodiment of the present invention.

As shown in FIG. 2, the influence evaluation system 100 according to an embodiment of the present invention includes a social data collection unit 110, an influence direction determination unit 140, a directional index calculation unit 150, and an intensity It includes an index calculation unit 160, and may further include a data analysis unit 120, a weight setting unit 130, and an output unit 170 together with these.

3 is a flowchart illustrating a method for evaluating interpersonal influence using social data according to an embodiment of the present invention.

The configuration and operation of the influence evaluation system 100 will be described in detail with reference to FIGS. 2 and 3. First, the social data collection unit 110 collects social data generated by social network activity behaviors of social network service users (S310). To this end, the social data collection unit 110 may include an SNS data collection module 112 that collects social data from a social network service providing server. The SNS data collection module 112 can access the resource of the corresponding service using the access token of the social network services to which the users are subscribed, and as a result, collect various social data generated on the social network. I can. In addition, as will be described again below, the social data collection unit 110 may further include a communication data collection module 114 that collects personal communication device usage data of social network service users from a communication company server or the like. The personal communication device usage data collected by the communication data collection module 114 is data generated by at least one of voice communication or text communication through a personal communication device, and includes sender and receiver information, communication attempt count information, and communication time information. And the like.

Then, the data analysis unit 120 analyzes the social data collected by the social data collection unit 110 and detects information to be used for interpersonal influence evaluation (S320). For example, the data analysis unit 120 may detect information on an action subject, an action type, an action time, an action number, a likelihood level, etc. related to a social network activity action. To this end, the data analysis unit 120 includes an action subject detection module 122 that detects each action subject information from social data, an action form detection module 124 that detects action type information, and an action that detects action time information. It may include a time detection module 126, an action number detection module 128 that detects action number information, a likeness level detection module 129 that detects likeness information indicated in the action content, and the like. In particular, the likelihood detection module 129 extracts keywords related to the sensibility of the author using text mining technology when there is a reaction action made of text, such as writing a comment, and based on these keywords, the sensibility detection module 129 By performing sentimental analysis, the author's preference is detected as a preset value.

Then, the weight setting unit 130 sets a weight to be assigned to each of the detected pieces of information. In this case, the weight setting unit 130 may set a predetermined value corresponding to each piece of information as a weight.

In FIG. 2, the data analysis unit 120 and the weight setting unit 130 are shown as independent components, but the influence direction determination unit 140, the directionality index calculation unit 150, and the intensity, which will be described later, according to an embodiment The necessary components may be allocated to each of the index calculation units 160 and configured to be integrated.

Then, the influence direction determination unit 140 determines the influence direction of the first user by using the action subject information and the action time information of the social network activity obtained from the collected social data (S340). The influence direction determination unit 140 may primarily determine the direction of influence by using the action type information. That is, when there is a post writing and a comment related thereto, the influence direction determination unit 140 may determine that the first user who writes the post has an influence on the second user who posts the comment. However, when the action type is the same or the direction of influence cannot be determined by the action type, such as when a comment is posted, the direction of influence can be determined using the action time information of each action. That is, the influence direction determination unit 140, when there is an action of a first user and an action of a second user related to each other, when the action of the second user follows the action of the first user, the influence of the first user It may be determined that the direction is toward the second user. The influence direction determination unit 140 may determine the direction of interpersonal influence using, for example, Equation 1.

[Equation 1]

In Equation 1, T _A denotes an action time of user A, T _B denotes an action time of user B, and ΔT denotes a time interval between both actions. Influence direction determination unit 140, when ΔT is greater than 0, that is, among the activities of user A and user B that are related to each other, when the activity of user B is a trailing activity, the direction of influence between user A and user B is It is judged to be heading to. In this case, it can be seen that the actions of user A are the causal actions and the actions of user B correspond to the reaction actions. On the other hand, when ΔT is less than 0, that is, when the user A's action is a subsequent action among both actions, the influence direction determination unit 140 determines that the influence direction between the two is from the user B to the user A. In addition, when ΔT is 0, the influence direction determination unit 140 may defer the determination of the direction of influence by considering both actions as independent.

Then, when it is determined that the direction of influence of the first user is toward the second user, the directional index calculation unit 150 includes a time interval between the causal action of the first user and the reaction action of the second user related to the causal action. In consideration of the directionality index for the direction of influence of the first user is calculated (S350). Here, the directional index corresponds to an index indicating the accuracy or reliability of determining the direction of influence. The directional index calculation unit 150 may calculate a directional index for the direction of influence of the first user, for example, using Equation 2.

[Equation 2]

In Equation 2, D _(A→B) is the directional index for the direction of influence from user A to user B, T _A is the cause action time of user A, T _B is the reaction action time of user B, and w is the corresponding reaction action. The weight of, N represents the number of reaction actions. In this way, the directional index calculation unit 150 assigns a weight (w) according to the form of the reaction behavior to the time interval between _{the first user's causal behavior (T A} ) and the second user's reaction behavior (T _{B ).} The directional index (D) may be calculated, and the directional index (D) may be calculated by further considering the number of reaction actions (N) of the second user. In addition, the directional index calculation unit 150 calculates and adds the directional index for each reaction behavior when _{the second user has the form (i) of the reaction behavior (T B) performed up to the m-th form.} You can calculate the directionality index. The directionality index (D) calculated as described above represents the accuracy or reliability of determining the directionality of the influence of the first user. In the case of calculating the directionality index as shown in Equation 2, the smaller the calculated directionality index value, the more accurate the determination of the influence direction by the influence direction determination unit 140 may be. That is, user B's response to user A's actions is fast, and as the number of times increases, the directionality index value decreases. As a result, it can be seen that the determination that the direction of influence of user A is toward user B is more reliable.

For example, for a post of user A registered at 10 am, user B made a comment at 10:30 and made a recommendation at 11:30, 12, and 15:00 respectively, and posted comments and recommended actions. When the weights (w ₁ , w ₂ ) for each of w ₁ =10 and w ₂ =1, it is determined that the direction of influence of user A is toward user B, and the directional index for the direction of influence is 5.9444 (= (10) ×(0.5)}/1 ² +{1×(1.5)+1×(2)+1×(5)}/3 ² ).

Meanwhile, the accuracy of the direction of influence indicated by the directional index may be determined as shown in Table 1.

Directional index (D) accuracy 0 to 10 Very high 10 to 100 height 100 to 200 usually 200 ~ 480 lowness 480 or more Very low

Therefore, it can be seen that the determination that the direction of influence of the user A is toward the user B has very high accuracy.

Meanwhile, as described above, when the social data collection unit 110 further collects personal communication device usage data of social network service users from a communication company server through the communication data collection module 114, the directional index calculation unit 150 ) May calculate the directional index by further considering communication direction information between the first user and the second user obtained from the collected personal communication device usage data. For example, the directionality index calculation unit 150, when a communication attempt corresponding to the response behavior of the second user occurs within a predetermined time after the occurrence of the cause behavior of the first user, that is, the sender is the second user and the receiver is the first When an attempt occurs in voice communication or text communication, which is a user, the accuracy of the directional index may be further increased by reducing the directional index value to a certain degree in consideration of the communication direction.

Then, when it is determined that the direction of influence of the first user is toward the second user, the intensity index calculation unit 160 includes information on the causal behavior of the first user and the reaction behavior of the second user related to the causal behavior. The intensity index for the intensity of the first user's influence is calculated using (S360). In this case, the intensity index calculation unit 160 uses only the information on the reaction actions of other users determined to be influenced by the user A according to the determination of the influence direction determination unit 140, and the intensity of the influence of the user A. You can calculate the index. First, the intensity index calculation unit 160 may calculate an influence intensity index of the first user on the second user by using Equation 3, for example.

[Equation 3]

In Equation 3, S _(A→B) is the index of the intensity of influence of user A on user B, ΔT is the time interval between the causal behavior of user A and half of the reaction behavior of user B, R is the likelihood of the reaction behavior, and N is The number of reaction actions, w _N , w _T , and w _R represent weights for N, T, and R, respectively. As such, the intensity index calculation unit 160 calculates the time interval (ΔT) between the causal behavior of the first user and the reaction behavior of the second user, the content of the reaction behavior (R), and the number of reaction actions (N). Considering the strength index (S) can be calculated. In addition, the intensity index calculation unit 160, the time interval (ΔT) between the causal action of the first user and the response action of the second user, the content of the reaction action (R), and the number of reaction actions (N), respectively. _{Weights (w N} , w _T , w _R ) according to the form of the reaction behavior By giving it, the strength index can be calculated. In addition, the intensity index calculation unit 160 calculates and sums the intensity index for each response activity when the form (i) of the response behavior performed by the second user exists up to the m-th form, The final influence intensity index of A can be calculated. In addition, the intensity index calculation unit 160 may calculate the total influence intensity index of the user A by calculating and summing the intensity indexes of the user A with respect to other users, which the influence of the user A has, respectively. The intensity index calculation unit 160 may calculate the total influence intensity index of user A using Equation 4, for example.

[Equation 4]

In Equation 4, S _A is the total influence intensity index of user A, S _(A→B) , S _(A→C) , S _(A→D) , S _(A→E) are users B and C, respectively User A's influence strength indexes, α _B , α _c , α _D , and α _E on, D, and E represent weights considering the influence of users B, C, D, and E, respectively. In this way, the intensity index calculation unit 160 may calculate the total influence intensity index (S _A ) of the user A by calculating and summing the intensity indexes of user A with respect to other users, respectively. In this case, the intensity index calculation unit 160 may calculate the total influence intensity index of user A by assigning weights considering the influence of other users to each of the intensity indexes of user A with respect to other users. That is, when user A exerts influence on other users having a high influence intensity index, the overall influence intensity index of user A may be higher.

Meanwhile, as described above, when the social data collection unit 110 further collects personal communication device usage data of social network service users from a communication company server through the communication data collection module 114, the intensity index calculation unit 160 ) May calculate the strength index by further considering information on the number of communication attempts between the first user and the second user obtained from the collected personal communication device usage data. For example, the intensity index calculation unit 160, when communication attempts corresponding to the reaction behavior of the second user occur within a predetermined time after the occurrence of the cause behavior of the first user, that is, the sender is the second user and the receiver is the first user. When human voice communication or text communication attempts occur, the accuracy of the strength index may be further increased by increasing the strength index value to a certain degree in consideration of the number of communication attempts.

Then, the output unit 170 outputs the result of evaluation of the influence of users of the social network service. In this case, the output unit 170 outputs the direction of the influence, the directional index, and the strength index calculated as a result of the influence evaluation as text, or outputs a graphic representing the influence between users in a time series using these calculated values. can do.

4 shows an example of an interpersonal influence evaluation result.

As shown in FIG. 4, the output unit 170 may output a graphic indicating a result of interpersonal influence evaluation in time series. That is, interpersonal influence rating of 4 results, the time with respect to the T _A user A has registered posts, time T _B is the user B Reply twice commented on, "like" input one time, and perform a one-time sharing, At time T _C , user C performs one comment, at time T _D , user D enters'Like' for user B's shared post once, and performs one share, and at time T _E , user E In the case of performing one input of'Like' on the shared post of user D, the direction and intensity of influence between users A, B, C, D, and E are analyzed in time series. In this case, the direction of the arrow may indicate the direction of influence, the length of the arrow may indicate the directionality index, and the thickness of the arrow may indicate the strength index of interpersonal influence. In an embodiment, the output unit 170 may display a directional index and an intensity index of interpersonal influence, and an overall influence intensity index of each user on a graphic.

Meanwhile, the embodiments of the present invention can be implemented by recording a computer program that can be read by a computer on a recording medium that can be read by a computer. When embodiments of the present invention are implemented using a computer program, the components of the present invention are program code segments that execute a necessary task. Such a computer program or code segments may be stored in a computer-readable recording medium or combined with a carrier wave and transmitted in the form of a data signal through a transmission medium or a communication network. Computer-readable recording media include all types of recording media that record data that can be read by a computer system. For example, the computer-readable recording medium may include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. In addition, program codes can be stored or executed in a distributed manner by distributing such recording media in computer systems connected through various networks.

As described above, according to the present invention, by evaluating influence using various social data generated by social network activity behaviors of social network service users, it is possible to evaluate interpersonal influence on various subjects, objects, and the like. In particular, it improves the accuracy and reliability of interpersonal influence evaluation by not simply measuring the intensity of influence, but by determining the direction of influence and calculating the influence intensity index in consideration of the determined influence direction. In addition, it is possible to verify the accuracy and reliability of the determination of the direction of influence by providing a numerical directional index for the determined direction of influence. In addition, by allowing time-series analysis of the direction and intensity of interpersonal influence in relation to a specific topic, target, etc., influence evaluation technology can be used in various ways such as market analysis, marketing strategy establishment, advertisement model selection, personal profile creation and verification It should be used for a variety of purposes in the field. Furthermore, it goes without saying that various embodiments according to the present invention can solve various technical problems other than those mentioned in the present specification in the related technical field as well as in the relevant technical field.

So far, the present invention has been described with reference to specific examples. However, those skilled in the art will be able to clearly understand that various modified embodiments may be implemented in the technical scope of the present invention. Therefore, the above-described embodiments should be considered from a descriptive point of view rather than a limiting point of view. That is, the scope of the true technical idea of the present invention is shown in the appended claims, and all differences within the scope of equivalents thereto are to be construed as being included in the present invention.

100: influence evaluation system 110: social data collection unit
120: data analysis unit 130: weight setting unit
140: influence direction determination unit 150: direction index calculation unit
160: strength index calculation unit 170: output unit

Claims (17)

A social data collection unit that collects social data generated by social network activities of social network service users;
An influence direction determination unit for determining an influence direction of one or more users by using information about an action subject and an action time of a social network activity obtained from the collected social data;
When it is determined that the direction of influence of the first user among the one or more users is toward the second user, the first user's cause action and the time interval between the reaction action of the second user related to the cause action are considered. 1 A directionality index calculation unit that calculates a directionality index for a user's influence direction;
When it is determined that the direction of influence of the first user is toward the second user, information on the cause action of the first user and the reaction action of the second user related to the cause action An intensity index calculation unit that calculates an intensity index related to the intensity of the influence; And
An output unit for outputting an interpersonal influence evaluation result including an influence direction of the first user, a direction index regarding the influence direction of the first user, and an intensity index regarding the influence strength of the first user; and
The influence direction determination unit,
When the first user's behavior and the second user's behavior related to the first user's behavior exist, and the first user's behavior and the second user's behavior are the same, the first user's behavior Based on the action type information of the action and the action of the second user, as well as the information on the action of the first user and the action of the second user, the direction of influence of the first user is 2 Interpersonal influence evaluation system using social data to determine whether or not the influence direction of the second user is toward the user or the influence direction of the second user is toward the first user. delete The method of claim 1,
The influence direction determining unit, when the second user's action related to the first user's action is later than the first user's action, determines that the influence direction of the first user is toward the second user. Interpersonal influence evaluation system using social data. The method of claim 1,
The directionality index calculation unit calculates the directionality index by assigning a weight according to the type of the reaction behavior to a time interval between the causal behavior of the first user and the reaction behavior of the second user. Interpersonal influence evaluation system. The method of claim 4,
The directionality index calculation unit calculates the directionality index by further considering the number of reaction actions of the second user. The method of claim 1,
The intensity index calculation unit calculates the intensity index in consideration of the time interval between the causal behavior of the first user and the reaction behavior of the second user, the content of the reaction behavior, and the number of reaction actions. Interpersonal influence evaluation system using social data. The method of claim 6,
The intensity index calculation unit assigns a weight according to the type of the reaction action to the time interval between the causal action of the first user and the reaction action of the second user, the content of the reaction action, and the number of reaction actions, respectively. Interpersonal influence evaluation system using social data, characterized in that calculating the intensity index. The method of claim 7,
The social data collection unit further collects personal communication device usage data generated by at least one of voice communication or text communication through personal communication devices of the social network service users,
The directional index calculation unit calculates the directional index by further considering communication direction information between the first user and the second user obtained from the collected personal communication device usage data,
The intensity index calculation unit calculates the intensity index by further considering information on the number of communication attempts between the first user and the second user obtained from the collected personal communication device usage data. Liver Impact Assessment System. In the method for the influence evaluation system interworking with the network server including the social network service providing server to evaluate the interpersonal influence using social data,
A social data collection step in which the influence evaluation system collects social data generated according to social network activity behaviors of social network service users;
An influence direction determination step of determining, by the influence evaluation system, an influence direction of one or more users by using information about an action subject of a social network activity and action time information obtained from the collected social data; And
When it is determined that the direction of influence of the first user among the one or more users is toward the second user, the influence evaluation system determines the time between the causal action of the first user and the reaction action of the second user related to the causal action A directional index calculation step of calculating a directional index for the direction of influence of the first user in consideration of an interval;
When it is determined that the direction of influence of the first user is toward the second user, the influence evaluation system uses information on the causal action of the first user and the reaction action of the second user related to the causal action. An intensity index calculation step of calculating an intensity index related to the intensity of the influence of the first user; And
Outputting, by the influence evaluation system, an interpersonal influence evaluation result including an influence direction of the first user, a directional index related to the influence direction of the first user, and an intensity index related to the influence strength of the first user; Including,
The determining of the direction of influence may include:
When the first user's behavior and the second user's behavior related to the first user's behavior exist, and the first user's behavior and the second user's behavior are the same, the first user's behavior Based on the action type information of the action and the action of the second user, as well as the information on the action of the first user and the action of the second user, the direction of influence of the first user is 2 determining whether or not the influence direction of the second user is toward the user or the influence direction of the second user is toward the first user; including
A method of assessing interpersonal impact using social data. delete The method of claim 9,
The step of determining whether the influence direction of the first user is toward the second user or the influence direction of the second user is toward the first user,
Social data, characterized in that, when the second user's behavior related to the first user's behavior follows the first user's behavior, determining that the direction of influence of the first user is toward the second user. Interpersonal influence evaluation method using. The method of claim 9,
The directionality index calculation step is a step of calculating the directionality index by assigning a weight according to the type of the reaction behavior to a time interval between the causal behavior of the first user and the reaction behavior of the second user. A method of assessing interpersonal impact using data. The method of claim 12,
The directionality index calculation step is a step of calculating the directionality index by further considering the number of reaction actions of the second user. The method of claim 9,
The intensity index calculation step is a step of calculating the intensity index in consideration of the time interval between the causal behavior of the first user and the reaction behavior of the second user, the content of the reaction behavior, and the number of reaction behaviors. A method of evaluating interpersonal influence using social data as a feature. The method of claim 14,
In the calculating of the intensity index, a weight according to the type of the reaction behavior is given to the time interval between the causal behavior of the first user and the reaction behavior of the second user, the content of the reaction behavior, and the number of reaction actions. The method for evaluating interpersonal influence using social data, characterized in that the step of calculating the intensity index. The method of claim 15,
The social data collection step is a step of further collecting personal communication device usage data generated by at least one of voice communication or text communication through personal communication devices of the social network service users,
The directionality index calculation step is a step of calculating the directionality index by further considering communication direction information between the first user and the second user obtained from the collected personal communication device usage data,
The step of calculating the strength index is a step of calculating the strength index by further considering information on the number of communication attempts between the first user and the second user obtained from the collected personal communication device usage data. Interpersonal influence evaluation method using. A computer program recorded on a recording medium as a computer program for executing the method according to any one of claims 9, 11 to 16 through a computer system.

Images